[CONTBIBUTION
NO. 192 FROM
THE
DEPARTMENT OF OBGANIC
CHEMISTRY, FORDHAM UNIVERSITY]
STUDIES ON THE CHEMISTRY OF HETEROCYCLICS. VIII. ENZYMATIC RESOLUTION OF RACEMIC @-2-THIENYLALANINE AND PREPARATION OF SOME SUBSTITUTED p-2-THIENYLALANINES' BERPU'ARD F. CROWEa AND F. F. NORD Received January 16, 1960
Previous studies in this laboratory have demonstrated the applicability of the N-methylformanilide (1) and rhodanine (2) syntheses to reactions in the thiophene series. Among the compounds made available through these investigations was the amino acid analog of phenylalanine, viz., 8-2-thienylalanine, which has proved to be of interest in growth inhibition experiments both on microorganisms (3) and on the rat (4). Since different effects were noted in these metabolism studies with the racemic amino acid and its optical isomers, it was decided to use an enzymatic method (4a) for its resolution thus obviating the more tedious chemical resolution previously applied (3). For this purpose ft carboxypeptidase preparation (5) was utilized because of its effectivenes in asymmetrically hydrolyzing chloroacetylphenylalanine (6). This investigation indicated some major differences between thienyl- and phenyl-alanine. It was found that while the enzyme did not attack chloroacetyl-D-thienylalanine,it completely hydrolyzed the L-isomer but only in the presence of a buffer. However no buffer was required for the enzymatic hydrolysis of chloroacetyl-L-phenylalanine(6). It n-as also noted that the chloroacetylated thienyl enantiomorphs were susceptible to racemization when hydrolyzed with 2 N HCI, followed by three successive evaporations to dryness. Substitution of 1 N HBr for 2 N HC1 with one evaporation produced optically-pure amino acids. The results from the two methods are compared in more detail in the experimental section. In view of the biological importance of p-2-thienylalanine the alkyl- and halogen-substituted thienylthiopyruvic acids, reported previously (2), were further reacted to obtain their corresponding thienyl amino acids. The preparation of the alkyl-substituted amino acids proceeded readily like that of the first member of the series, but the halogen-substituted acids presented some unforeseen results. When the 5-bromothienyloximino acid was reduced with 2% sodium amalgam, hydrogenolysis occurred yielding 0-2-thienylalanine instead of the expected p-(5-bromo-2-thienyl)alanine(Chart I). This was proved conclusively by the analysis and melting point of the amino acid isolated and from the melting point of its phenylurea derivative. Prior investigations have demonstrated the removal of bromine from the thiophene ring with hydrogen in the presence of palladium on charcoal (7, 8). 1 For Papers V I and VI1 of this series see J . Org. Chem., 16,81,89 (1950). This investigation was carried out under the auspices of the Office of Naval Research. For a preliminary communication see Arch. Biochem., 26, 460 (1950). The analyses were carried out by Dr. F. Buhler and A. A. Sirotenko of this Department. * Abridged from a part of the dissertation presented t o the Graduate School of Fordham University, 1950, in partial fulfillment of the requirements of the degree of Doctor of Philosophy. 688
689
CHEMISTRY O F HETEROCYCLICS. VI11
The preparation of p- (5-chloro-24hienyl)alaninewas achieved in lower yields than the alkyl-substituted products and under anomalous conditions. In the other cases cited above, the free amino acid precipitated from solution after the reaction mixture had been cooled. The chloro compound failed to precipitate even after standing for one week in a refrigerator. However when the solution was diluted with water t o twice the original volume and chilled, the amino acid CHARTI
I/
s
NOH
"2
I: I
PH FIGURE1. TITRATION &RVES ALANINE.
Water blank
alanine --.-.--.
-
OF fl-2-THIENYLALAS1K.E AND
B-(B-CHLORO-2-THIENYL)-
, 8-2-Thienylalanine - - - -; p-(5-Chloro-2-thienyl)-
appeared. It may be noted that since small additions of lactic acid were used to keep the reduction media acidic, the pH of the solution at the end of the reaction was about 5.7,while the pH of the solution after dilution was 4.9, A marked difference between p-2-thienylalanine and ~-(5-chloro-2-thienyl)alaninewas shown by titrating them in an aqueous HCI solution with NaOH. These results are presented in Figure 1. These observations indicate that the presence of the chlorine atom has an appreciable effect on the properties of the resultant amino acid.
690
BERNARD F. CROWE .4ND F. F. NORD
EXPERIMENTAL'
ChZoroacetyZ-p-2-thienyZ-nL-aZanine. Applying the method described for the chloroacetylwas dissolved in ation of L-phenylalanine (9), 9.5 g. (0.055 mole) of @-2-thienyl-~~-alanine 56 cc. of 1 N NaOH and placed in an ice-bath. To this solution was added, with shaking and in small portions, 83 cc. of 1 ilr NaOH and a solution of 8.8 g. of chloroacetyl chloride in ether. Dry hydrogen chloride gas was then passed into the alkaline mixture for 30 minutes, which precipitated the chloroacetylated compound as an oil. After standing overnight cold, the oil solidified giving a yield of 11.35 g. (82%). Recrystallization from hot water (Norit) gave 9.75 g. of white crystals (717,). Carboxypeptidase preparation. The enzyme was obtained from 10 kilos of fresh-frozen beef pancreas (5), and stored with 480 cc. of water at -14'. One-eighth of this suspension was used for each enzymatic digestion. The amount of chloroacetylated compound used as substrate varied between 9.5 g. and 30 g.
SPECIFIC ROTdTIONS
OF
TABLE I OPTICAL ISOMERS
I
8-2-Thienylalanine Chloroacetyl-~-2-thienylalanine
..
jI 1
P-2-THIENYLALANINE I
[.I
I
comoum
OF
[.I LITERATURE
OBSERVED
L-form
-31.4= $46.5"
I
i
,
D-form
1
L-form
I
D-form
I
+31.4* -47.2d
1
-31.7
-
(12)
f31.6 I
-
(12)
1.0834 g. in 50 cc. of water. b 0.2624 g. in 25 cc. of water. ~0.8468g. in 50 cc. of absolute alcohol. d0.9487 g. in 50 cc. of absolute alcohol. Resolution procedure. Chloroacetyl-p-2-thienyl-~~-alanine (9.5 g., 0.038 mole) was suspended in water and dissolved by the cautious addition of 6 N LiOH with vigorous stirring. The enzyme was brought into solution by adjusting the suspension to p H 7.6 with 0.2 M LiOH. After the insoluble globulins were removed by filtration, 75 cc. of MacIlvaine buffer (pH 7.6) was added and the solution was mixed with the substrate, forming a total volume of about 300 cc. The digestion was carried out a t 37' for 48 hours. The mixture was then acidified with glacial acetic acid t o p H 5.0, and evaporated in vacuo to half of its original volume. The L-amino acid filtered off at this point amounted t o 1.2 g. (36%) after washing with cold water and absolute alcohol. The filtrate was concentrated further in vacuo, layered with ethyl acetate, acidified to p H 2.0 with conc'd HCI, and extracted five times with ethyl acetate. The extract was dried with sodium sulfate and evaporated i n vacuo to an oil which crystallized after washing with petroleum ether. Upon filtration there was This product, after recrystalobtained 3.2 g. (67%) of chloroacetyl-p-2-thienyl-~-alanine. lization from hot water, had a specific rotation of [a]: -47.2', which is in agreement with its L-antipode obtained by chloroacetylation of the L-amino acid. Chloroacetyl-@-2-thienyl-~-alanine (1.35 g., 0.055 mole) was refluxed for 70 minutes with 1 N HBr and evaporated to dryness i n vacuo. The white residue was dissolved in absolute alcohol and adjusted to pH 5.2 with 4N LiOH whereupon 0.65 g. (69%) of the D-amino acid precipitated. I n a previous experiment no buffer was used during the enzymatic digestion and while some optically-pure L-amino acid was obtained, the chloroacetylated D-compound was evidently contaminated with unhydrolyzed chloroacetylated L-amino acid because i t had the specific rotation [a]:-31.1'. This mixture when hydrolyzed with 2 N HC1 yielded a free amino acid with no optical activity. 3
The authors wish to acknowledge the cooperation of Dr. J. V. Fiore and S. N. Timasheff.
69 1
CHEMISTRY OF HETEROCYCLICS. VI11
Data relative t o the four optically-active compounds described are recorded in Table I. To verify the identity of the chloroacetylated racemic acid, used as the starting material in the resolution, i t was converted t o the unsaturated azlactone by treatment with acetic anhydride and pyridine (10). This product showed no depression in melting point when TABLE I1 ANALYTICAL AND MELTING-POINT DATAOF OPTICALISOMERS COMPOUND
I
M.P.,
m=BoGEN
Calc'd
Found
Observed
p-2-Thienyl-~~-alanine
8.18
8.27
p-2-Thienyl -D -alanine
8.18
8.38
p-2-Thienyl-~-alanine
8.18
8.25
Chloroacetyl-j%2-thienyl - ~ ~ - a l anine Chloroacetyl-p-2-thienyl-~-alanine Chloroacetyl-P-2-thienyl-~-alanine 2-Methyl4-(2-thenal)-5-oxazolone~ 2-Methyl-4-(2-thenaI)-5-oxazolone*
5.65
5.85
273-275' 238-244d 273-276c 239-246d 272-274~ 238-244d 127-128
5.65
5.90
119.5-120
5.65
5.65
120.5-121.5
7.26 7.26
7.34 7.27
131-132.5 131.5-132.5
'C.
1
From chloroacetyl-~-2-thienyl-~~-alanine. T,From 2-thenaldehyde. bath to 270". Oil-bath heated a t rate of 8'/minute.
Literature
274-275" (13) 243-245d (3) 239-244d (12) 239-244 (12)
c
Preheated oil-
TABLE I11 5-THIENYL-SUBSTITUTED 3-(~-THIENYL)-2-OXIMINOPROPIONICACIDSDERIVED THROUGH THE RHODANINE SYNTHESIS
5-THIENYL SUBSTITUENT
Y.P."'
1, OC.
YIELDb,
%
Calc'd
Found
C Ii c -__________
5-Methylc . . . . . . . . . . . . . . . . . . . 155-156 5-Ethyl'. .................... 152.5-153.5 5-Propyld. . . . . . . . . . . . . . . . . . . . 143-144 5-Chloroe. . . . . . . . . . . . . . . . . . . . 153.5-154.5 5-Bromo6.. . . . . . . . . . . . . . . . . . . 156-157
94 94 93 95 90
48.24 50.70 52.90 38.26 31.81
4.52 5.16 5.72 2.73 2.27
48.05 50.90 53.20 38.29 31.78
Ii
4.44 5.23 5.33 2.73 2.44
Oil-bath heated rapidly t o within 7-8' of the recorded melting point. b Yields represent per cent conversion from the corresponding 2-thienylthiopyruvic acids. c Recrystallized from toluene. d Recrystallized from benzene. * Recrystallized from chloroform. 1 All melting points indicate decomposition. 0
mixed with an authentic sample prepared from 2-thenaldehyde by means of the Erlenmeyer azlactone synthesis (11). The analyses and melting points of the isomeric amino acids and their chloroacetylated compounds are listed in Table I1 together with those of the azlactone. Preparation of oximino propionic acids. The procedure employed was the same for all, as exemplified by the preparation of the 5-methyl compound. Benzene and chloroform were
BERNARD F. CROWE AND F. F. NORD
I I
=tl d V
I ssi I
zl
y:
lp-
c
In
CHEMISTRY OF HETEROCYCLICS. VIII
693
substituted for toluene in the recrystallization of some of these products. These variations are indicated in Table I11 as well as other pertinent data relative to the compounds thus synthesized. 3-(5-Methyl-2-thienyl)-2-oximinopropionic acid. To a solution of 1.7 g. of sodium in 49 cc. of ethanol was added 5.16 g. of hydroxylamine hydrochloride. This solution after filtraacid and heated on tion was added t o 4.6 g. (0.023 mole) of 5-methyl-2-thienylthiopyruvic a steam-bath for 30 minutes. After removal of the alcohol under a vacuum the residue was dissolved in 12 cc. of 5% NaOH and filtered. The cooled filtrate was carefully acidified under ether with 11 cc. of 10% HCI and the precipitated acid filtered off and dried in vacuo over potassium hydroxide. The water layer was removed in a separatory-funnel and extracted three times with ether. The combined ether extracts were dried over Drierite and evaporated to dryness. Total yield, 4.31 g. (94%). Preparation of amino acids and their ureides. The preparation of &(5-methyl-2-thienyl)alanine described below was followed in all instances with the exception that the 5-chloro compound was precipitated by diluting the final reaction mixture with water to twice the original volume. All the amino acids prepared exhibited a positive ninhydrin reaction. The phenylurea derivatives were prepared in the usual manner with phenyl isocyanate. In Table IV are listed the compounds so obtained. B-(5-Methyl-2-thienyl)alanine.3-(5-Methy1-2-thienyl)-%-olriminopropionicacid (4.41 g., 0.022 mole) was dissolved in 60 cc. of absolute alcohol and 200 g. of 2% sodium amalgam was added in small portions with heating on a steam-bath. Small additions of lactic acid were made during the reduction to keep the solution acidic to litmus. When addition of the the amalgam was complete, the solution was decanted and left in the refrigerator overnight. Filtration afforded 2.85 g. of amino acid. By evaporation of the solution t o a syrup and treatment with absolute alcohol a further 0.2 g. was obtained; total yield, 3.05 g. (74%). The aqueous solutions of 0-2-thienylalanine and @-(5-chloro-2-thienyl)alanine used for titration contained 0.2424 g. and 0.4443 g. of amino acid per 100 ml. of solution, respectively. A Fisher titrimeter was employed to obtain the data presented in Figure 1. SUMMARY
Racemic P-2-thienylalanine was resolved into its optical isomers through the asymmetric hydrolysis of its chloroacetylated derivative using a purified beef pancreas carboxypeptidase preparation. Four thienyl amino acids were prepared from 5-methyl-2-thena1dehyde2 5-ethyl-Zthenaldehyde, 5-propyl-Z-thenaIdehyde,and 5-chloro-Z-thenaldehyde. NEW YORK58, N. Y. REFERENCES (1) KINGAND Nom, J . Org. Chem., 13,635 (1948). (2) CROWEAND N o m , Nature, 163, 876 (1949); CROWELVD NORD,J . Org. Chem., 16, 81 (1950). (3) DU VIGNEAUD, et al., J . Biol. Chem., 169, 392 (1945.) (4) FERGER AND DU VIGNEAUD, J . Biol. Chem., 179, 61 (1949). (4a) NEUBERG AND LINHARDT, Biochem. Z., 147,372 (1924). (5) ANSON, J . Gen. Physiol., 20, 663 (1937). (6) GILBERT,PRICE, AND GREENSTEIN, J . Biol. Chem., 180,473 (1949). (7) MOZINGO, et al., J . Am. Chem. SOC.,67, 2092 (1945). (8) DITTMER,J . Am. Chem. Soc., 71, 1205 (1949). (9) FISCHER AND SCHOELLER, Ann., 367, 1 (1907). (10) BERGMA", ZERVAS, AND LEBRECHT, Ber., 64, 2315 (1931). (11) HERBSTILVD SHEMIN,Org. Syntheses, Coll. Vol. 11, 1 (1943). (12) FERGER AND DU VIGNEAUD, J . Biol. Chem., 174, 241 (1948). (13) BARGERAND EASSON, J . Chem. SOC.,2102 (1938).
